Wikipedia talk:Wikipedia Signpost/2017-06-23/News and notes
Discuss this story
- Wikipedia:List of administrators states 1262 admins (544 active, changes in Wikipedia talk:List of administrators). User:NoSeptember/Admin_stats lists 1463 in 2012 (latest date).
Extrapolating a linear loss at last five year rate gives (which doesn't take into account the current zero new admins rate):annual loss at (1463-1262)/5 = 40.2 admin/yearno admins in (1262/40.2) = 31 years, the year 2048
active annual loss at (744-529)/5 = 43 active/yearno active admins in (529/43) = 12 years, the year 2029
- Is "Departments reorganized at Wikimedia Foundation, and a month without new RfAs (so far)" meant to be "wikt:rearrange the deck chairs on the Titanic"? Widefox; talk 12:29, 23 June 2017 (UTC)
- You can't use a linear extrapolation. The number of active administrators is proportional to a Poisson point process and is a subset of active editors, which have stabilized over the past five years. 153.120.214.254 (talk) 10:50, 24 June 2017 (UTC)
- Linear is simplistic, what fits better? The annual losses of active admins since 2008 are: 42, 86, 88, 33, 81, 30, 50, 1, 53 (a mean of 51.5). A simple linear regression of active totals since 2008: r2=0.96, standard error = 32, 50 loss a year., extrapolation to 0 takes 9 years, the year 2026. An exponential decay model visually fits (unlike linear), and r2=0.98, the trend halving the active admins per decade, henceforth "active admin attrition (AAA) rate" (so 1000 ten years ago, 500 now, predicts 250 in 2027).Widefox; talk 09:59, 26 June 2017 (UTC)
- Here's a graph with better analysis (exponential regression fits only from mid 2011, total has r2=0.989 but a linear is similar) predicts less active admins than semi-active in a year from now, and total admin numbers 50% down from the peak by 2027:
- Linear is simplistic, what fits better? The annual losses of active admins since 2008 are: 42, 86, 88, 33, 81, 30, 50, 1, 53 (a mean of 51.5). A simple linear regression of active totals since 2008: r2=0.96, standard error = 32, 50 loss a year., extrapolation to 0 takes 9 years, the year 2026. An exponential decay model visually fits (unlike linear), and r2=0.98, the trend halving the active admins per decade, henceforth "active admin attrition (AAA) rate" (so 1000 ten years ago, 500 now, predicts 250 in 2027).Widefox; talk 09:59, 26 June 2017 (UTC)
- You can't use a linear extrapolation. The number of active administrators is proportional to a Poisson point process and is a subset of active editors, which have stabilized over the past five years. 153.120.214.254 (talk) 10:50, 24 June 2017 (UTC)
-
User:Widefox/editors English Wikipedia administrator numbers 2003-2019, total number (and % from peak), % active, % semi-active, % inactive -
User:Widefox/editors English Wikipedia administrator numbers 2011-2019 (ditto legend) -
User:Widefox/editors English Wikipedia administrator numbers 2011-2029 (ditto legend)
Widefox; talk 21:37, 30 June 2017 (UTC)
- What are "audience verticals"? Not all of us use such jargon. Thanks. Julietdeltalima (talk) 19:29, 23 June 2017 (UTC)
- Vertical market probably, and yes, the announcement is very heavy on Silicon Valley biz jargon. - Bri (talk) 21:26, 23 June 2017 (UTC)
- It wasn't entirely clear in the announcement and there was no time to get clarification before the publication deadline, so I thought it would be best to stick to their lingo. I'd like the record to show that I did try to figure it out though!!! —A L T E R C A R I ✍ 10:08, 24 June 2017 (UTC)
- The term "audience vertical" was introduced in the 2015 re-org. The idea is that job functions (like coding or testing) are horizontal slices through an organization, while products and features are vertical slices. In the foundation, each "vertical" focuses on a specific audience (e.g. Readers), and at least in theory each one contains personnel of all the disciplines it needs to be relatively self-sufficient. Each "vertical" is smaller than a department, but usually larger than a team. KSmith (WMF) (talk) 18:42, 26 June 2017 (UTC)
- Vertical market probably, and yes, the announcement is very heavy on Silicon Valley biz jargon. - Bri (talk) 21:26, 23 June 2017 (UTC)
- I know I'm confused. As I recall, the community re-elected Raystorm and Pundit as well as returned Doc James to the seat he was wrongly removed from. I can see that WMF keeps several editors in non-voting positions (no surprise) but I guess I don't understand what Doc's position means. Chris Troutman (talk) 10:09, 24 June 2017 (UTC
- I was wondering and asked Doc James in this secion of his user talk page. The !election for the WMF Board only provides advice / recommendations to the Board, which controls the appointment of its own members. The meeting where the decision will be taken is not until August. Apparently, if Doc James is appointed to the Board, he would also move from a non-voting / advisory position on the Committee to which he has been appointed to become a full voting member of it. He sees this advisory appointment as a positive sign for the actions of the Board in appointing its new members in August, so hopefully this (temporary) advisory appointment is good news. EdChem (talk) 14:20, 24 June 2017 (UTC)
- "We trained hard but it seemed that every time we were beginning to form into teams, we would be reorganized. I was to learn later in life that we tend to meet any new situation by reorganizing. And what a wonderful method it can be for creating the illusion of progress while producing inefficiency, confusion and demoralization." -- Attributed to Gaius Petronius Arbiter (c.27-66 AD).